Re: Dispersion Test-whether Poisson or NB?

Posted by Rara on May 02, 2014; 3:29pm
URL: http://spssx-discussion.165.s1.nabble.com/Dispersion-Test-whether-Poisson-or-NB-tp5725769p5725775.html

Thank you very much for your response. From the likelihood ratio test, I think it's better to use the NB. I also read from this link (IBM SPSS Guide) that if I set the parameter of NB to 0 and check the logrange multiplier test, that would show the over-dispersion as well. I got these results and I don't know how to interpret them. Does this mean that the test is significant and the over-dispersion exist for this dataset?

Lagrange Multiplier Test:
                                           z Parameter < 0   Parameter > 0 Non-directional
--------------------------------------------------------------------------------------------------------
Ancillary Parameter 133.333     1.000                  .000                 .000
       
Thanks,
Rara